A three-dimensional (3D) model of an object may describe various aspects of the object (e.g., dimensions, color, texture). With sufficient amounts of data, 3D models may be used in various 3D printing applications, interactive visualizations, measurements, and many other suitable applications. However, collection and generation of data for the creation of 3D models may be complex (e.g., for non-rigid objects such as body parts). Furthermore, capture and generation of 3D models may require sophisticated and expensive equipment such as various sensors. Additionally, capture and generation of 3D models can require significant amounts of time and/or computational resources.
In some examples, orthotics and prosthetics require accurate models of human extremities including the geometry of the object to be modeled. The automatic modeling of human beings or particular body parts (e.g., faces, arms, hands, feet, legs) may be challenging because the objects may not be rigid or there may be challenges associated with holding still for periods of time (e.g., children, the elderly, or trembling patients). Additional challenges may relate to the natural skin of body parts not necessarily being well textured for 3D reconstruction.